[BugFix] Fix torch distributed stateless PG backend init (#14870)

Signed-off-by: Nick Hill <nhill@redhat.com>
2025-03-15 20:26:19 -07:00
parent 71c1e07107
commit b82662d952
2 changed files with 8 additions and 3 deletions
--- a/examples/offline_inference/data_parallel.py
+++ b/examples/offline_inference/data_parallel.py
@ -76,5 +76,10 @@ if __name__ == "__main__":
                             GPUs_per_dp_rank))
        proc.start()
        procs.append(proc)
+    exit_code = 0
    for proc in procs:
        proc.join()
+        if proc.exitcode:
+            exit_code = proc.exitcode
+
+    exit(exit_code)